Efficient Refrigerator DIY
- Thread starter Trebor English
- Start date
Help Support Van Living Forum:
scottorious
Well-known member
- Joined
- Oct 26, 2017
- Messages
- 64
- Reaction score
- 0
Very interesting read, Would there be any benefit to using copper pipe in place of PVC? I can only come up with them being more space conscious as the walls are thinner and the endcaps not so bulky. I guess the same heat transfer would happen in both just slightly slower in PVC? Maybe the copper pipe would be beneficial on the freezing stage????
This is deep into the weeds and the physics is a bit beyond me but would a DIY polar tube with a slight vacuum drawn on it create a situation where the water freezes at a slightly higher temp but as it "thaws" it sublimate directly to gas? Would that phase change be even more endothermic than the phase change from solid to liquid? I suppose that only a limited amount of sublimation would occur before the pressure reached a point inside the tube that just normal melting would occur.... just a crazy late night though.
This is deep into the weeds and the physics is a bit beyond me but would a DIY polar tube with a slight vacuum drawn on it create a situation where the water freezes at a slightly higher temp but as it "thaws" it sublimate directly to gas? Would that phase change be even more endothermic than the phase change from solid to liquid? I suppose that only a limited amount of sublimation would occur before the pressure reached a point inside the tube that just normal melting would occur.... just a crazy late night though.
scottorious
Well-known member
- Joined
- Oct 26, 2017
- Messages
- 64
- Reaction score
- 0
This looks like a cool product that could be used to make any size ice pack and already has ratios provided for those of us who don't want to try and recall 9th grade chemistry class.
https://koolergel.com/
Would be cool if the OP replaced his baggy of water with koolergel and tested setup again. Hint Hint!
https://koolergel.com/
Would be cool if the OP replaced his baggy of water with koolergel and tested setup again. Hint Hint!
Yes copper would transfer heat better coming and going, but super pricey.
Would be interesting to compare meltpoints of that "gel" mix with seawater.
And raises the fact seawater can be diluted, it does contain a LOT of salt.
If, you have your little freezer with you, no need to waste so much storage space on massive long-lasting sizes, just rotate 1" ones daily.
Also may allow running the freeze cycle only during sun hours with panels going.
Would be interesting to compare meltpoints of that "gel" mix with seawater.
And raises the fact seawater can be diluted, it does contain a LOT of salt.
If, you have your little freezer with you, no need to waste so much storage space on massive long-lasting sizes, just rotate 1" ones daily.
Also may allow running the freeze cycle only during sun hours with panels going.
scottorious
Well-known member
- Joined
- Oct 26, 2017
- Messages
- 64
- Reaction score
- 0
It looks like lowest freezing point that you can reach with salt dissolved in water would be -6 degrees F. What I found online is that koolergel says it will freeze between 0 and -10 degrees F. So unless there is another positive to the "gel" it might be easiest to just make up a salt water solution of 27% salt by weight.
That's a **lot** saltier than seawater.
You do actually want your solution to freeze, maximum eutectic energy absorption happens at the phase change from solid to liquid.
If you make it too salty, some freezers won't go down that low, especially in hot ambient.
For example some don't get below +10°F in 90° weather.
The ideal solution would have a freezing point that is a bit higher than the worst case for **your** freezer.
You do actually want your solution to freeze, maximum eutectic energy absorption happens at the phase change from solid to liquid.
If you make it too salty, some freezers won't go down that low, especially in hot ambient.
For example some don't get below +10°F in 90° weather.
The ideal solution would have a freezing point that is a bit higher than the worst case for **your** freezer.
Trebor English
Well-known member
scottorious said:Would be cool if the OP replaced his baggy of water with koolergel and tested setup again. Hint Hint!
With the Koolergel, like with salt, the freezing point can be made very low. I want a fridge. If I am keeping lettuce and a tomato I don't want it frozen. Just the regular freezing point of water is cold enough. A few degrees colder wouldn't freeze produce but it would not be an advantage.
When 1 pound of water melts it absorbs 144 BTUs. From what I can tell with the help of Google, adding salt doesn't change that. With salt it does the absorbing at a lower temperature but it is still 144 BTUs per pound. That would be ignoring the weight of the salt. If you add .27 pounds of salt to 1 pound of water the mix would be 1.27 pounds. If the references Google found are correct that 1.27 pounds would still absorb 144 BTUs when melting.
Since I have spilled once I am convinced that it could happen again. I expect that the idiot will be more careful in the future. I am still a slow learner but in this case I know that salt water is bad to spill in a steel van and it can be avoided by not using it.
Trebor English
Well-known member
Flexible polyethylene bags make good thermal contact with the bottom of the freezer compartment. They conform to the irregular surface of the refrigerant tubes. There is a large area that conducts the heat out of the water. Rigid containers like polar tubes don't have that. The larger diameter caps on the ends will keep the middle part of the tube elevated above the bottom of the freezer compartment. The air gap will insulate. The spots where the end caps touch the evaporator tubes have a very small area.
With a low thermal conductivity between the freezing water in a polar tube and the refrigerant the temperature of the refrigerant will be colder. That makes a higher pressure differential in the compressor resulting in higher consumption of electricity. Even with the colder freezer compartment the heat in the water will not flow from the water to the freon nearly as quickly without conduction. Convection through air is not as effective as conduction. The run time of the compressor will be longer resulting in higher consumption of electricity.
Thermal conductivity, k, is reciprocal of thermal resistance, r-value, k = 1/r
401 Copper
205 Aluminum
43 Steel, Carbon 1%
2.18 Ice (0C, 32F)
0.58 Water
0.33 Polyethylene, LDPE
0.19 PVC
0.04 Glass wool insulation
0.024 Air
Normally a cup of water in the freezer will freeze at the top first. With the poly bags in contact with the evaporator tubes of the freezer compartment ice forms first at the bottom. Initially there is no ice, eventually there is 2 inches so on average there is an inch of ice. The thin layer, 2 mils (.002 inch) of poly (k=.33) and then an inch of ice (k=2.18) will conduct the heat from the freezing boundary between the ice and water to the freon quite effectively.
The key to making this fridge efficient is minimizing the compressor run time. The compressor runs at about 3500 RPM due to the AC being 60 Hz, 3600 cycles per minute. Reducing the RPM is how the 12 volt fridges get low current use, A 35 Hz inverter would make the RPM about 2000. The inverter I got for $20 is 60 Hz. There is no magic 270 ohm resistor on the circuit board to remove to slow it down. Good thermal conductivity and 5 pounds of water lets the compressor pump out a day's worth of BTUs in 1 to 2 hours.
With a low thermal conductivity between the freezing water in a polar tube and the refrigerant the temperature of the refrigerant will be colder. That makes a higher pressure differential in the compressor resulting in higher consumption of electricity. Even with the colder freezer compartment the heat in the water will not flow from the water to the freon nearly as quickly without conduction. Convection through air is not as effective as conduction. The run time of the compressor will be longer resulting in higher consumption of electricity.
Thermal conductivity, k, is reciprocal of thermal resistance, r-value, k = 1/r
401 Copper
205 Aluminum
43 Steel, Carbon 1%
2.18 Ice (0C, 32F)
0.58 Water
0.33 Polyethylene, LDPE
0.19 PVC
0.04 Glass wool insulation
0.024 Air
Normally a cup of water in the freezer will freeze at the top first. With the poly bags in contact with the evaporator tubes of the freezer compartment ice forms first at the bottom. Initially there is no ice, eventually there is 2 inches so on average there is an inch of ice. The thin layer, 2 mils (.002 inch) of poly (k=.33) and then an inch of ice (k=2.18) will conduct the heat from the freezing boundary between the ice and water to the freon quite effectively.
The key to making this fridge efficient is minimizing the compressor run time. The compressor runs at about 3500 RPM due to the AC being 60 Hz, 3600 cycles per minute. Reducing the RPM is how the 12 volt fridges get low current use, A 35 Hz inverter would make the RPM about 2000. The inverter I got for $20 is 60 Hz. There is no magic 270 ohm resistor on the circuit board to remove to slow it down. Good thermal conductivity and 5 pounds of water lets the compressor pump out a day's worth of BTUs in 1 to 2 hours.
Yes I agree I would not carry salt water if there was a high risk of spillage.
The purpose of the lower temp is the longer "lifespan" of the tube/block before the cooler gets to say the 45° swapping point.
As I said even after the melting point the tube itself will remain below 32° for quite a long time.
Perhaps the ideal is to have a mix of tubes with different strength solutions, so the phase change events keep happening at different times, maybe even days apart.
The purpose of the lower temp is the longer "lifespan" of the tube/block before the cooler gets to say the 45° swapping point.
As I said even after the melting point the tube itself will remain below 32° for quite a long time.
Perhaps the ideal is to have a mix of tubes with different strength solutions, so the phase change events keep happening at different times, maybe even days apart.
Trebor English
Well-known member
I presume that by rotate you don't mean turn them 90 degrees but rather take them out of the freezer and put them in the fridge. I only have one cold box. The Haier 1.7 cubic foot fridge is it. The water bags go in the freezer compartment of the fridge.John61CT said:
Trebor English
Well-known member
I now have new Ziploc bags of water, no salt. I wanted to put 10 pounds of water in it. I tried 3 bags with 3.3 pounds in each. I tore 2 before I got them in. Then I tried 5 bags of 2 pounds each. I tore the fifth one. I have decided to settle for 8 pounds of ice.
With that much water in the freezer compartment there is no room for food on top of the ice as there was before. With no food stored there there will be no spilling into the top of the ice area. There will be no need to pull out the ice bags to clean it up. These bags should last a long time without being abused.
The first compressor run took 8.5 hours. I used a shore power plug in to get it all frozen in one shot. Since the water wasn't cold from a prior compressor run the performance is what I expected. With 8 pounds it freezes about 1 pound per hour after the water gets cooled to close to freezing.
The testing I did before was when the weather was still hot. Any testing now with the temperature 20 degrees colder is of much less interest to me so I probably won't bother doing any till it gets close to summer. I expect that the performance will be the same as before. Electricity use will be 7 to 12 amp hours per day, less in winter. Melting will continue to be 2.5 pounds per day, less in winter. Now there is more ice so it will last more days, 3, without any compressor running.
I had a thought. Batteries don't like partial state of charge cycling. If you recharge to only 80% then discharge some more the lead sulfate that is more deeply buried gets harder to get rid of by charging. You need more voltage and more time. The more this happens the harder it is to fully recharge and then battery capacity starts to drop. By doing the energy storage with ice this doesn't happen. With my small solar system it happens often that after a couple of rainy days the water doesn't get completely frozen in one day. The water doesn't have a diminished capacity after partial cycling.
The first bag that failed was a store brand. The bags that I tried to use with 3.3 pounds of water were store brand. For the 5 bags of 2 pounds I bought Ziploc brand. Four out of 5 had zipper leaks that I fixed with Dicor as before. I was not surprised. When putting them in the freezer compartment it was much easier with only 2 pounds of water per bag. Like all the store brand bags that failed, the tear that leaked was a tiny scratch about a quarter inch long with a pinhole at one spot. The name brand doesn't appear to be an advantage. I did find some thick bags, 8 mil, on the internet but a full case of 500 bags is more than I want to spend.
With that much water in the freezer compartment there is no room for food on top of the ice as there was before. With no food stored there there will be no spilling into the top of the ice area. There will be no need to pull out the ice bags to clean it up. These bags should last a long time without being abused.
The first compressor run took 8.5 hours. I used a shore power plug in to get it all frozen in one shot. Since the water wasn't cold from a prior compressor run the performance is what I expected. With 8 pounds it freezes about 1 pound per hour after the water gets cooled to close to freezing.
The testing I did before was when the weather was still hot. Any testing now with the temperature 20 degrees colder is of much less interest to me so I probably won't bother doing any till it gets close to summer. I expect that the performance will be the same as before. Electricity use will be 7 to 12 amp hours per day, less in winter. Melting will continue to be 2.5 pounds per day, less in winter. Now there is more ice so it will last more days, 3, without any compressor running.
I had a thought. Batteries don't like partial state of charge cycling. If you recharge to only 80% then discharge some more the lead sulfate that is more deeply buried gets harder to get rid of by charging. You need more voltage and more time. The more this happens the harder it is to fully recharge and then battery capacity starts to drop. By doing the energy storage with ice this doesn't happen. With my small solar system it happens often that after a couple of rainy days the water doesn't get completely frozen in one day. The water doesn't have a diminished capacity after partial cycling.
The first bag that failed was a store brand. The bags that I tried to use with 3.3 pounds of water were store brand. For the 5 bags of 2 pounds I bought Ziploc brand. Four out of 5 had zipper leaks that I fixed with Dicor as before. I was not surprised. When putting them in the freezer compartment it was much easier with only 2 pounds of water per bag. Like all the store brand bags that failed, the tear that leaked was a tiny scratch about a quarter inch long with a pinhole at one spot. The name brand doesn't appear to be an advantage. I did find some thick bags, 8 mil, on the internet but a full case of 500 bags is more than I want to spend.
Trebor English
Well-known member
Now that the space above the ice is not available for food I have recalculated. The space for food is now 13.7 quarts, 12.9 liters, 787 cubic inches. The freezer compartment is 395 cubic inches. It isn't full of water but there is so much in there that I can't get more in.
Trebor English
Well-known member
I said lower ambient temperature testing was less interesting and I probably wouldn't do it. Being geeky I couldn't resist.
After 2 days running to stabilize (8.5 hours day 1, 44 minutes day 2) the next 4 days averaged 9.12 amp hours per day. The average running current is lower probably because the temperature in the van is lower. Today it is cloudy so there will be no compressor time.
After 2 days running to stabilize (8.5 hours day 1, 44 minutes day 2) the next 4 days averaged 9.12 amp hours per day. The average running current is lower probably because the temperature in the van is lower. Today it is cloudy so there will be no compressor time.
Headache
Well-known member
Trebor how is this going?
I just built a cooler box for my 32 quart cooler and now a 10 lb block of ice lasts almost 10 days. I'm in my cooler a minimum of 4x a day. I'm really looking forward to when I can have a fridge/freezer.
I just built a cooler box for my 32 quart cooler and now a 10 lb block of ice lasts almost 10 days. I'm in my cooler a minimum of 4x a day. I'm really looking forward to when I can have a fridge/freezer.
Trebor English
Well-known member
Headache said:
It works. The cold space is tiny. Only one 100 watt panel is barely enough. The 8 pounds of ice really does work. It holds 2 or 3 days with no electricity but then it takes a couple of days to re-freeze all the water again. Two 100 watt panels would make it so that almost every day would be a good solar day but I don't have space on the roof of my minivan. I had the 100 watt panel, the $20 inverter and the 75 amp hour starter battery. The fridge was $80, the thermostat $5, and the added insulation about $20.
I give it 5 stars for cheapness. The outside bulk gets zero stars. It is big outside for the 12.9 liter inside size. I also give it zero stars for the effort to make it work. It is an interesting science project.
Amazon sells a small compressor fridge, Alpicool C15, 15 liters, $200, and Alpicool, C20, 20 liters, $240. Either ought to work with 200 watts solar and a pair of golf cart batteries. From the reviews it looks like 144 watt hours per day, 12 amp hours per day, set at refrigerator temps. It might hold temperature well enough to run it during the day and off at night.
ArmorAbby had a C15 but New Jersey in the winter was not a good test of it. It was below 32 degrees inside the van.
skyl4rk
Well-known member
- Joined
- Nov 30, 2014
- Messages
- 681
- Reaction score
- 14
If you are looking for a way to allow the compressor to run only when there is a good charge from the solar panels, here is an idea:
Install a small solar panel not connected to any battery, and connect the output of that panel to the voltage sensor of the Low Voltage Disconnect. The small solar panel should be mounted near your main solar panels so it receives the same amount of sun. The solar panel output might need a resistor in the circuit to prevent turning things on under low light conditions, for example, parked under streetlights. Experimentation to determine the appropriate shut off/turn on settings would be needed.
When the separate solar panel is in sunshine and putting out power, the refrigeration system turns on. The thermostat allows it to run until ice is frozen, then it shuts the inverter off and the battery bank is charged. When the separate solar panel no longer receives enough sunlight, its output voltage drops and it disconnects the refrigeration system from the battery bank. The system then uses chilled ice to stay cool until the next sunny day.
A switch could be used to connect battery bank power to the Low Voltage Disconnect voltage sensor to allow the user to turn the fridge on manually, running from the battery bank power only.
Perhaps also connect the voltage sensor to a solenoid that turns on voltage when the alternator is putting out power.
Install a small solar panel not connected to any battery, and connect the output of that panel to the voltage sensor of the Low Voltage Disconnect. The small solar panel should be mounted near your main solar panels so it receives the same amount of sun. The solar panel output might need a resistor in the circuit to prevent turning things on under low light conditions, for example, parked under streetlights. Experimentation to determine the appropriate shut off/turn on settings would be needed.
When the separate solar panel is in sunshine and putting out power, the refrigeration system turns on. The thermostat allows it to run until ice is frozen, then it shuts the inverter off and the battery bank is charged. When the separate solar panel no longer receives enough sunlight, its output voltage drops and it disconnects the refrigeration system from the battery bank. The system then uses chilled ice to stay cool until the next sunny day.
A switch could be used to connect battery bank power to the Low Voltage Disconnect voltage sensor to allow the user to turn the fridge on manually, running from the battery bank power only.
Perhaps also connect the voltage sensor to a solenoid that turns on voltage when the alternator is putting out power.
Headache
Well-known member
I haven't looked lately but there was a guy in Prescott selling 310 watt panels for $150. Jim in Denver told me about it. You'd have to buy an mppt controller for it. That's not a bad deal if you're in the area and if he's still selling.
frater secessus
Well-known member
Right now there is a used one listed on Amazon for $99, free shipping.Trebor English said:
https://www.amazon.com/gp/offer-listing/B073WYS3TR/ref=dp_olp_all_mbc?ie=UTF8&condition=all
